On Tue, Apr 19, 2011 at 01:15:11AM +0200, Ingo Schwarze wrote:
> Hi Matthew and Joachim,
> 
> Matthew Dempsky wrote on Mon, Apr 18, 2011 at 11:34:50AM -0700:
> 
> > The issue here (if any) is that we over-specify the *successful*
> > return value as precisely 0, rather than generally non-negative.
> 
> I like the general idea, so i'd suggest the following.
> Note that the exact wording has been chosen to be as close
> to the output of the .Rv macro as possible.
> 
> In case we go into that direction, i suspect that some other
> manuals might contain similar overspecifications.
> 
> Yours,
>   Ingo
> 
> 
> Index: uname.3
> ===================================================================
> RCS file: /cvs/src/lib/libc/gen/uname.3,v
> retrieving revision 1.12
> diff -u -r1.12 uname.3
> --- uname.3   31 May 2007 19:19:29 -0000      1.12
> +++ uname.3   18 Apr 2011 23:05:00 -0000
> @@ -63,11 +63,14 @@
>  Machine hardware platform.
>  .El
>  .Sh RETURN VALUES
> -If
> +The
>  .Fn uname
> -is successful, 0 is returned; otherwise, \-1 is returned and
> +function returns a non-negative value (on
> +.Ox ,
> +always 0) if successful; otherwise the value -1 is returned
> +and the global variable
>  .Va errno
> -is set appropriately.
> +is set to indicate the error.
>  .Sh ERRORS
>  The
>  .Fn uname
> 

Why not just use the POSIX statement replacing 'shall' with 'is'?
"Upon successful completion, a non-negative value is returned.
Otherwise, -1 is returned and errno is set to indicate the error."

Why create hostages to fortune by being over-specific?

.... Ken

Reply via email to